This is a mysterious codec that i recently find out about it , codec architect apparently is very different from h.264 or h.265 and vp9.
I couldn't play demo sample after installing codec i hope some of you have success with it .
description of SIF1
Description of SIF1 settings
SIF1 video compression format specification
+ Reply to Thread
Results 1 to 30 of 39
Well, I ran UniExtract on the pesky installer, and then registered the damn .ax file
Anyway, the demo clips didn't impress me at all
And AVI is the only container supported thus far
IMHO, that codec is somewhere between beta and alpha, still too far from being useful.
1. SIF transform core code restructurization and multithreaded optimization.
2. Adding a new higher-quality SIF transform modes through the use of PsyRD optimization.
3. Adding support for quarter-pixel motion compensation.
4. Writing a portable SIF1 encoding-decoding library for Linux.
5. Performing SSE2 and multiprocessor optimization of the current code.
6. Further development of new algorithms based on SIF-technology.
I tried it once as VfW codec. And was not impressed (Sintel Trailer was hardly recognizable). But that may have been an early version. Though ... according to still missing features, no. I doubt there was any further development since.
The developer did everything alone
Do not have enough money for the development of the project and had to bury.
His mistake was that he contacted the American monopoly licensing system - spent a lot of money and left with nothing. It was necessary to apply to the Chinese...
This is developer last comment two years ago .
On the last some months I didn't work over the project.
Unfortunately I have no possibility to work over the project now.
For all the time of existence of the project I didn't have any serious support from anybody.
Therefore I do not think that local guys seriously worry about destiny of this project.
In too time the project isn't closed.
I will work over it when I have free time.
I worked over it twelve years and I am going to work over it and in the future.
But it is necessary to understand that I am not Google.
I can't employ ten developers and invest in the project the necessary millions for development.
Therefore and further the project will development with such speed.
Thanks for understanding.
God forbid, but if it dies of a heart attack or a car crash, then what his family would do with this patent?
- It will still have to share with someone or sell the project.
BTW, it turns out that the Elecard is a representative of DivX in Russia
A youtube video? If you can't host a video yourself and use a video service which always reencodes that is simply bad PR.
Looking at the youtube video the sf1 part looks better in some frames and worse in others, but without encoding and decoding speeds that doesn't make it interesting at all,...
-> another PR fail
Nowadays you need:
a. support in libav (at least for decoding) since more and more commercial tools use it + it gives you cross plattform usage
b. support for linux/mac/windows
c. up-to-date news page&co (speed and video quality comparisons,...)
Last edited by Selur; 19th Sep 2019 at 23:14.users currently on my ignore list: deadrats, Stears555
Powered by GoDaddy GoCentral Website Builderusers on my IgnoreList: 162 names thus far, but featuring DB83 and manono.
Sorry, I think I have replied to previous comment, but looks like it did not pass through (may be because a link to our full detailed presentation). So, I will try again:
1. Web site: PR is not our highest priority now - we ahve made it as temporarily landing. I think, professional version will come closer to CES and / or NAB
2. Our top priority is to reach h265+ level
3. I will try to post our presentation with more detail, but next reply to be here text information will pass. We are going through international codec competition each years - results will be availble soon
4. We plan to release Libav support as well as mobile plug-in for Androis and iOS pretty soon (Oct - Nov)
5. Yes, we support Win and Linux now
I could not even discover yet what the acronym "SIF" is supposed to mean. Neither the documentation page nor the patent file explain that.
At least I find it interesting that you may have developed another promising kind of decomposing and re-composing video frames. I'll stay curious.
The links to the SIF samples and codec in the presentation are 404, another sign of poor professionalism.
I'm not sure what you're hoping to achieve by posting about your very early work on this, but fair warning, many of the posters here have a great amount of knowledge about video and a very critical eye. You've already been called on your use of Youtube and your the links to your samples and codec are mysteriously broken. If you want to be taken seriously, think twice and double check everything before your next post. It appears that English is not your native language, so be extra careful about what you post.
Most of the posters here, especially the regulars are very friendly and helpful, but draw the line at puffery and outright false claims. A couple of years ago, someone from another company tried to build up support here and was highly criticized for failing to provide the quality promised in his samples (at the time compared to H.264). The final straw was that he never provided the software for posters to perform their own controlled tests as he promised. Not only was he chased out of this forum for his false claims and broken promises, and to date, AFAIK, his revolutionary codec never got any support from anyone else.
[QUOTE=lingyi;2560664]The links to the SIF samples and codec in the presentation are 404, another sign of poor professionalism.
Do you mean in presentation or on old web site? Sorry about it - I will correct it asap
Here's the thread from three years ago: https://forum.videohelp.com/threads/380675-Announcing-ZPEG-Demonstration-Site-Works-wi...-h265%21/page3
A prime example of what to expect if you plan to continue to post about your product here and how NOT to conduct yourself.
Sorry, guys! It may be sharing issue - will provide all corrected links within 24 hours.
As promised latest and correct info and links:
SIF Codec version 1.78
All files with same setting are compresses in one step. Bit rate is controlled by embedded Human Vision Model (HVM). "Frame to frame" compression mode was used - no predictors at all
Mobile quality (Bit rate 999Kb 0.027 bit per pixel):
Regular quality (Bit rate 2.198Mb 0.060 bit per pixel):
Same by h265 (Bit rate 2.142Mb 0.058 bit per pixel):
2. STAR WARS
Mobile quality (Bit rate 1.389Mb 0.038 bit per pixel):
Regular quality (Bit rate 2.866Mb 0.078 bit per pixel):
Same by h265 (Bit rate 3.071Mb 0.083 bit per pixel):
3. ROGUE ONE
Mobile quality (Bit rate 1.254Mb 0.034 bit per pixel):
Regular quality (Bit rate 3.077Mb 0.084 bit per pixel):
Same by h265 (Bit rate 2.945Mb 0.080 bit per pixel):
Not the best choice of acronyms: SIF has long (<=3 decades) referred to "source input format" a constrained parameter version of MPEG1 & 2. Equal to VCD quality.
People who remember are going to think this is just another mediocre quality mpeg1 vfw codec.
If it's not better than AV1, open source, and free -- there's no need and no point.
Is there an encoder component included ? I can find registered x86/x64 directshow decoder
But the Sif1_vfw.dll does not seem registered , and I cannot register manually (some problem with the dll)
In SysWOW64 there are Sif1_vfw.dll, Sif1Dec.ax
System32 has Sif1Dec_64.ax
Not to be a kill joy, who am I kidding, I love being a kill joy, the links this guy provided lead to full versions of the listed movies, encoded in x265 (I downloaded the ones that say "Same by h265"), not to sample encodes comparing his encoder to x264/x265.
Technically, in some countries, this would be considered "pirated" content and against this forum's rules.
OP: Thanks for the downloads, if you feel like posted some more "pirated" content...
Console versions (same compression ration, much higher speed)